Suggested next steps to prevent cervical cancer after surgical treatment for high-grade cervical dysplasia

The risk of cervical cancer among women treated for high-grade cervical dysplasia is more than twofold compared with the general population, and this risk remains elevated for over two decades. In Sweden, cervical cancer incidence is rising again and the risk of cervical cancer among women with a prior history of high-grade cervical dysplasia has increased since the 1960s. The surgical procedure known as conization is commonly used to treat high-grade cervical dysplasia and prevent progression to invasive cervical cancer. However, treatment failure, defined as residual/recurrent/ high-grade cervical dysplasia or cervical cancer post-conization, has reportedly increased by almost twenty percent. Suggested risk-factors for post-conization treatment failure include age, smoking, treatment modality, lesion size and severity, incomplete excision of lesion, infection with high-risk human papilloma virus (hrHPV) and hrHPV persistence. The overarching aim of this thesis addresses how to protect women from developing cervical cancer following treatment of high-grade cervical dysplasia. The included studies examine risk factors for recurrent disease and what factors or combinations thereof can accurately predict treatment failure and thereby identify women at high risk post-conization.

Study I investigated the long-term risk of residual/recurrent high-grade cervical dysplasia post-conization and how such risk varies according to margin status, comorbidity and HPV infection. The study included a total of 991 women who had undergone conization for high-grade cervical dysplasia between 2000 and 2007. Data were obtained from medical records and the Swedish National Cervical Screening Registry (NKCx). Given a median follow-up of ten years and maximum of sixteen years, almost twelve percent of the cohort was diagnosed with residual/recurrent disease or worse (invasive cervical cancer). A greater than 2.5-fold risk of recurrent disease was found among women with incomplete resection compared with cases where the margins were clear. Risk varied according to the extent of anatomical infiltration of disease margins and was particularly elevated when endocervical margins were positive. Comorbidities such as autoimmune disease, HIV, hepatitis B and/or C, malignancy, diabetes, and genetic disorder and/or organ transplantation were independent predictors of recurrent disease. For the subgroup of women who were hrHPV positive with involved margins, risk of recurrent disease was increased compared with the subgroup of women who were HPV positive with clear margins. Women with incompletely resected precancerous lesions are at increased risk for recurrent/residual high-grade cervical dysplasia and cervical cancer. Combined assessment of margin and hrHPV status, while also taking comorbidities into account, may provide a useful strategy to accurately identify at-risk women who should undergo reconization.

Study II evaluated risk of recurrent disease among women who had undergone first-time treatment for high-grade cervical dysplasia, within a cohort where complete HPV status was known. A total of 529 women were included, all of whom had undergone conization for high-grade cervical dysplasia between 2014 and 2017. Follow-up continued for up to six years post-conization, during which time 22 patients were diagnosed with recurrence of high-grade cervical dysplasia. Four significant independent risk factors for recurrence were identified: age 45 or older, involved margins, positive hrHPV test at first follow-up and abnormal cytology at first follow-up. Furthermore, persistent hrHPV infection was associated with recurrent disease. The finding that involved margins are an independent risk factor suggests that more intense follow-up is required for these women, regardless of early HPV status post-conization. Although early HPV-positive status post-treatment was found to be a strong independent risk factor for predicting recurrent disease, more than 30% of the 22 patients diagnosed with recurrent disease were HPV-negative shortly after treatment. These patients, however, were subsequently found to be HPV-positive on routine screening, suggesting that repeated HPV testing is necessary during post-conization follow-up.

Study III explored risk factors for recurrent/persistent adenocarcinoma-in-situ (AIS), as well as risk factors for progression from AIS to invasive cervical cancer among women who had previously undergone conization for AIS. A total of 84 women who had primary treatment with conization for AIS between 2001 and 2017 were included. Twelve women developed recurrent disease, two of whom had invasive cervical cancer. Among all factors, one or more positive HR-HPV assays post-conization provided the highest sensitivity for predicting recurrence, while smoking or past history of smoking were associated with the highest specificity for recurrence. When adjusting for age at conization and abnormal cytology at follow-up, we demonstrated that HPV18 positive status was the strongest predictor for post-conization recurrence. Two or more positive HPV results post-conization helped predict recurrence. The strong predictive value of HPV in relation to recurrence, especially HPV18, indicates that HPV testing during post-treatment follow-up for AIS is necessary. In addition, it is important to consider smoking status and to encourage long-term follow-up so as to better protect these women who are at high risk of recurrence and progression to invasive cervical cancer.

In conclusion, this thesis improves our understanding of what risk factors are able to accurately predict treatment failure and how to identify women at risk of recurrent disease after treatment. This thesis highlights the importance of individualized long-term follow-up, including evaluation of margin status based on residual tumor classification, the need for repeated HPV testing during follow-up and attention to comorbidities.

List of scientific papers

I. Alder S, MEGYESSI D, Sundström K, Östensson E, Mints M, Belkić K, Arbyn M, Andersson S. Incomplete excision of cervical intraepithelial neoplasia as a predictor of the risk of recurrent disease-a 16-year follow-up study. Am J Obstet Gynecol. 2020 Feb;222(2):172.e1-172.e12.
https://doi.org/10.1016/j.ajog.2019.08.042

II. Andersson S*, MEGYESSI D*, Belkić K, Alder S, Östensson E, Mints M. Age, margin status, high-risk human papillomavirus and cytology independently predict recurrent high-grade cervical intraepithelial neoplasia up to 6 years after treatment. Oncol Lett. 2021 Sep;22(3):684. *Contributed equally.
https://doi.org/10.3892/ol.2021.12945

III. Belkić K, Andersson S, Alder S, Mints M, MEGYESSI D. Predictors of treatment failure for adenocarcinoma in situ of the uterine cervix: Up to 14 years of recorded follow-up. Oncol Lett. 2022 Aug 25;24(4):357.
https://doi.org/10.3892/ol.2022.13477